JPH11355580A - Mtf correcting circuit and image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an MTF correcting circuit and an image processor with which a sharp image can be outputted even when image data are inputted without sufficiently following the resolution or amplitude of an input image. SOLUTION: Filters (a) and (b) have different characteristics and respectively output enhancement levels L1 and L2 for enhancing the high frequency component of image data. A maximum value detecting part 32 detects the maximum value of enhancement levels L1 and L2 outputted from the filters (a) and (b). Tables (a) and (b) output enhancement coefficients coefa and coefb corresponding to the density of a concerned pixel. A selector 34 selects either the enhancement coefficient coefa or the enhancement coefficient coefb corresponding to the codes of the enhancement levels L1 and L2. A multiplier circuit 35 multiplies the enhancement level L from the maximum value detecting part 32 and the enhancement coefficient coef outputted from the selector 34. An adding circuit 36 adds a density D0 of the concerned pixel and the output value of the multiplier circuit 35.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an MTF correction circuit and an image processing apparatus applied to a read image processing portion such as a facsimile machine, a digital copying machine, and a scanner.

[0002]

2. Description of the Related Art In a facsimile apparatus, a digital copying apparatus, a scanner apparatus, and the like, various processes are performed on an image input from an image reading portion in order to improve image reproducibility. For example, an MTF correction process for emphasizing a spatial high-frequency component of an image to raise a character edge portion is performed. However, when spatial high-frequency components are enhanced by MTF correction, high-frequency noise components are also enhanced. Therefore, in order to prevent image quality deterioration due to high frequency noise components, it is conceivable not to apply edge emphasis so much. However, in such a case, fine characters and lines may not be clearly reproduced.

In order to prevent such inconvenience, Japanese Patent Application Laid-Open No. 7-66997 discloses an edge-enhanced image in which a fixed Laplacian filter is applied to image data for each pixel of a predetermined matrix. There is disclosed a method of suppressing an amplitude component of a level not present in image data before edge enhancement in data. However, even with such a method, if an image reading device such as a scanner cannot follow the resolution and density amplitude of the input image,
Since edge enhancement cannot be performed and resolution cannot be sufficiently obtained, a blurred image is output.

Japanese Patent Application Laid-Open No. 7-245709 discloses an output (edge amount) of a digital filter for edge enhancement.
Is converted according to the sign of the edge amount to obtain an enhancement amount. The method is configured to weaken the MTF correction when the edge amount is negative, that is, when the target pixel is located at a position where the density changes from high density to low density in the image, in order to prevent white spots. I have. Then, when the spatial frequency of the image becomes high and the image reading device cannot follow the resolution and the density amplitude of the input image, the MTF correction of the white background part becomes weak, and the blurred image is reproduced again. I will.

[0005]

As described above, according to the conventional image processing method, when an image reading device such as a scanner cannot follow the resolution and density amplitude of an input image, effective edge enhancement is performed. There is a problem that a blurred image whose resolution is not sufficient cannot be reproduced.

Accordingly, an object of the present invention is to provide an MTF correction circuit and an image processing apparatus which can output a clear image even when image data which cannot sufficiently follow the resolution and amplitude of an input image is input. And

[0007]

An MTF correction circuit according to the present invention includes a filter section for calculating an enhancement amount based on a density difference between a target pixel and peripheral pixels of input image data, and an output of the density and filter section of the target pixel. Output the enhancement coefficient corresponding to the sign of the enhancement amount to be performed. When the density of the pixel of interest is at an intermediate level, when the enhancement amount is negative, a value larger than that when the enhancement amount is positive is obtained. An enhancement coefficient output unit that outputs an enhancement coefficient, a multiplication circuit that multiplies the enhancement amount output by the filter unit with an enhancement coefficient output by the enhancement coefficient output unit, and a multiplication value obtained by the multiplication circuit and the density of the pixel of interest are added. And an adder circuit for outputting the corrected density.

The filter section may include a plurality of filters having different characteristics, and a maximum value detecting section for outputting the maximum value of the outputs of the filters as an enhancement amount. The filter unit includes two filters. One filter is a filter having a coefficient for enhancing a density difference between a target pixel and a peripheral pixel positioned at an angle of 45 degrees with respect to the target pixel, and the other filter is a filter having a coefficient for enhancing the density difference with respect to the target pixel. Alternatively, the filter may have a coefficient that emphasizes a density difference between peripheral pixels on the front, rear, left, and right sides. The enhancement coefficient output unit includes a first table corresponding to a case where the sign of the enhancement amount is positive or zero, a second table corresponding to a case where the sign of the enhancement amount is negative, and a second table corresponding to the sign of the enhancement amount. And a selector for selecting any one of the output of the first table and the output of the second table. Further, in the first table and the second table, a large value coefficient may be set for the target pixel of low density.

An image processing apparatus according to the present invention reads an image and outputs digital image data, a digital image data output from the image reading apparatus including a line including a target pixel, and a digital image corresponding to several lines before and after the line. A line buffer for holding data; a filter unit for calculating an enhancement amount based on a density difference between a target pixel and peripheral pixels of the digital image data; enhancement according to the density of the target pixel and the sign of the enhancement amount output by the filter unit An enhancement coefficient output unit that outputs a coefficient, and outputs a larger enhancement coefficient when the enhancement amount is negative than when the enhancement amount is positive when the density of the target pixel is at an intermediate level, A multiplying circuit for multiplying the emphasis amount output from the filter unit by the emphasis coefficient output from the emphasis coefficient output unit; Those having a MTF correction circuit including an adding circuit for outputting a corrected concentration by adding.

The filter section in the MTF correction circuit may be configured to include a plurality of filters having different characteristics and a maximum value detecting section for outputting the maximum value of the outputs of the filters as an enhancement amount.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a transmission unit of a facsimile apparatus which is an example of an image processing apparatus according to the present invention. In the figure, a scanner 10 reads a document, converts read image data into digital image data by an internal A / D converter, and outputs the digital image data to a line buffer 20. The line buffer 20 holds image data from the scanner 10. The MTF correction circuit 30 performs MTF correction for each pixel. Also, γ
The correction unit 40 performs γ correction on the MTF-corrected image data and outputs the result to the binarization unit 50. The binarization unit 50 binarizes the image data and stores the binarized image data in the image memory 60.

The printer 70 prints out the binarized data in the image memory 60. Further, the communication unit 80 controls to send the binarized data in the image memory 60 to the line.

In the MTF correction in this embodiment,
The density of the target pixel is D0 and the density of the corrected target pixel is D0.
1. If the output (enhancement amount) of a digital filter (generally a Laplacian filter) is L and the enhancement coefficient is coef, D1 = D0 + coef × L. In this embodiment, as described below, a more appropriate value can be set as the emphasis amount L, and the emphasis coefficient coef can be set to a more appropriate value for outputting a clear image.

FIG. 2 is a block diagram showing the configuration of the MTF correction circuit 30. Filter a31a, filter b31
b has different characteristics, and outputs enhancement amounts L1 and L2 for enhancing the high-frequency components of the image data. The maximum value detector 32 includes a plurality of filters a31a and a plurality of filters b
The maximum value of the emphasis amounts L1 and L2 output by 31b is detected, and the maximum value is output as the emphasis amount L. Further, the sign of the enhancement amount L is determined. Table a33a and table b33b are
Enhancement coefficients coefa, coef according to the density of the pixel of interest
b is output.

The selector 34 has an emphasis coefficient coefa or an emphasis coefficient coe according to the judgment result of the maximum value detecting section 32.
fb is selected and output to the multiplication circuit 35 as the enhancement coefficient coef. The multiplication circuit 35 multiplies the enhancement amount L from the maximum value detection unit 32 by the enhancement coefficient coef output from the selector 34. Then, the adding circuit 36 calculates the density D of the target pixel.
0 is added to the output value of the multiplication circuit 35 to generate a correction value D1.

Next, the operation will be described. Scanner 1
0 reads the original contents and outputs digital image data to the line buffer 20. The line buffer 20
A line including the pixel of interest and image data for several lines before and after the line are held.

In this embodiment, a Laplacian filter shown in FIG. 3 is used as the filter a31a. The Laplacian filter shown in FIG. 4 is used as the filter b31b. FIG. 5 is an explanatory diagram illustrating a relationship between a target pixel and peripheral pixels. In FIG. 5, D
It is assumed that 0 is the density of the target pixel, and D10 to D33 are the densities of the peripheral pixels. Then, the enhancement amount L1 output from the filter a31a is expressed as follows using the filter coefficients shown in FIG.

The enhancement amount L2 output from the filter b31b is expressed as follows by using the filter coefficients shown in FIG.

The filter coefficients in the Laplacian filter are effective in improving the blur caused by a low MTF. However, when filtering is strongly applied, even a small change in the halftone portion of the image is emphasized, and the image becomes rough.
On the other hand, if it is weakly applied, sharpness in the character portion cannot be obtained.
In this embodiment, the coefficient of the filter a31a is a coefficient that emphasizes the density difference between the target pixel and the peripheral pixels located at an angle of 45 degrees, and the coefficient of the filter b31b is the front, rear, left, and right with respect to the target pixel. This is a coefficient for emphasizing the density difference from the surrounding pixels. The maximum value detection unit 32 includes a filter a31a
And the maximum value of the enhancement amounts L1 and L2 output by the filter b31b is selected and output as the enhancement amount L.

Therefore, when there is a density difference in an oblique direction in the image, the target pixel is emphasized by the filter a31a, and when there is a density difference in the front, rear, left and right directions, the target pixel is emphasized by the filter b31b. . Therefore, the target pixel can be effectively emphasized regardless of any density change without increasing the filter coefficient.

In this embodiment, the enhancement coefficient coef is determined according to the sign of the enhancement amount L and according to the density of the target pixel D0. Specifically, table a3
3a and the table b33b output a coefficient corresponding to the density D0 of the target pixel. The selector 34 selects the output of the table a33a when the enhancement amount L is positive or zero, and selects the output of the table b3 when the enhancement amount L is negative.
3b is selected, and the selected value is output to the multiplication circuit 36 as the enhancement coefficient coef.

FIG. 6 is an explanatory diagram showing an example of the configuration of the table a33a. Here, it is assumed that the digital image data is represented by 8 bits, 00 (H) indicates the maximum value on the white side, and FF (H) indicates the maximum value on the black side. The table a33a is used when the enhancement amount L is positive or zero, that is, when the density D0 of the target pixel changes to a higher density (black side) than the surrounding pixels. FIG. 7 shows the table b
It is explanatory drawing which shows an example of a structure of 33b. Table b3
3b is used when the enhancement amount L is negative, that is, when the density D0 of the target pixel changes to a lower density (white side) than the surrounding pixels.

The table a33a and the table b
Each coefficient of 33b is determined according to characteristics of the scanner 10 or the like to be used. In this embodiment, the following values are adopted as an example. coef1>coef2>coef4> coef3 coef5>coef6> coef7 coef1 = coef5 = 3 coef2 = coef6 = 2 coef3 = 0.5 coef4 = coef7 = 1 However, these values are examples and are actually used. An appropriate value is set according to the characteristics of the scanner 10 and the like.

Next, referring to the flowchart of FIG.
The operation of the TF correction circuit 30 will be described in more detail. The maximum value detection unit 32 includes a filter a31a and a filter b3.
The absolute values of the enhancement amounts L1 and L2 output by 1b are compared.
When | L1 | ≧ | L2 |, L1 is output as L1 | L1 | <| L2 | (steps S1 to S3).

As described above, in this embodiment, a plurality of filters having different characteristics are used in consideration of the fact that one filter characteristic may cause a peripheral pixel in which the density difference from the target pixel D0 cannot be detected.

When the enhancement amount L is positive or zero, the maximum value detection section 32 sets the select signal S supplied to the selector 34 to 0 (steps S4 and S5). When the enhancement amount L is negative, the select signal S is set to 1
(Steps S4 and S7). The selector 34 determines that S =
In the case of 0, the enhancement coefficient c output from the table a33a
oefa is selected as the enhancement coefficient coef. In this embodiment, the enhancement coefficient coefa is one of coef1 to coef4 corresponding to the density D0 of the target pixel. When S = 1, the selector 34 selects the enhancement coefficient coefb output from the table b33b as the enhancement coefficient coef. In this embodiment, the emphasis coefficient coefb is set to a value corresponding to the density D0 of the target pixel.
ef5 to coef7.

As an example, when the enhancement amount L is positive and the density of the target pixel D0 is 40 (H), the selector 34 selects coef3 output from the table a33a. That is, coef3 is input to the multiplication circuit 35 as the enhancement coefficient coef. Then, the multiplication circuit 35
Is the enhancement amount L and the enhancement coefficient c output by the maximum value detection unit 32.
oef and outputs the multiplied value to the addition circuit 36.

The addition circuit 36 adds the multiplied value output from the multiplication circuit 35 to the density D0 of the target pixel. Adder circuit 36
Is the density D1 of the target pixel after MTF correction.

As described above, in this embodiment, the MT
The F correction circuit 30 uses different emphasis coefficient tables depending on the sign of the emphasis amount L. That is, different types of enhancement coefficient tables can be used when the density D0 of the target pixel changes to a higher density side and to a lower density side than the surrounding pixels. Therefore, an appropriate emphasis coefficient can be set according to the density status of the target pixel and the peripheral pixels.

For example, when the spatial frequency of an image is high, sufficient density is obtained when the pixel of interest is a pixel on a black line, but the density decreases when the pixel of interest is on a white background. In other words, there is a possibility that a black level is determined when binarized. Therefore, in this embodiment, FIG.
As shown in FIG. 7 and FIG.
(H), the enhancement coefficient coef7 when the enhancement amount L is negative (when the density D0 of the target pixel changes from the black side to the white side), and the enhancement amount L is positive or zero. In this case (when the density D0 of the target pixel changes from the white side to the black side), the emphasis coefficient coef3 is set larger.

When the density is at an intermediate level and the enhancement amount L is negative, the value of [-coef × L] becomes larger (in absolute value), and the value of the corrected density D1 becomes 00.
(H) side closer. In other words, it is emphasized on the white side. As a result, the resolution of the black line increases. As described above, when the spatial frequency of the input image is increased and the scanner 10 cannot follow the resolution and the density amplitude of the input image and the level of the white background cannot be completely reduced, the white back is emphasized to the white side. In addition, black and white can be clearly reproduced without the black line being emphasized too much and the white back surface is not blackened. In other words, even when the spatial frequency of the input image becomes high and the scanner 10 cannot follow the resolution and the density amplitude of the input image, a clear image can be obtained according to the MTF correction of this embodiment. it can.

Further, as shown in FIGS. 6 and 7, the emphasis coefficients coef1, coef2, coef5, and coef6 used when the density D0 of the target pixel is low are increased, so that a thin character such as a pencil character is used. The manuscript can be reproduced clearly.

In this embodiment, a plurality of filters having different characteristics are used. However, even when one filter is used, two types of emphasis coefficient tables are switched and used according to the sign of the filter output. Thereby, the effect of improving the image quality when the spatial frequency of the input image is high is exhibited.

[0034]

As described above, according to the present invention, the MTF
The correction circuit and the image processing device output an enhancement coefficient corresponding to the density of the target pixel and the sign of the enhancement amount output from the filter unit. When the density of the target pixel is at an intermediate level, the enhancement amount is reduced. Since the image processing apparatus includes the enhancement coefficient output unit that outputs an enhancement coefficient having a larger value when the amount of enhancement is negative than when the amount of enhancement is positive, the spatial frequency of the input image is increased, and the image reading apparatus may perform If the level of the white background cannot be reduced due to the inability to follow the resolution or density amplitude, the white background is emphasized more to the white side, and the black line is emphasized too much so that the white background is not blackened. Can be clearly reproduced, and a clear image can be obtained.

In the case where the filter unit is configured to include a plurality of filters having different characteristics and a maximum value detection unit that outputs the maximum value of the outputs of the filters as an enhancement amount, the filter coefficient Can be effectively emphasized even if there are various density changes without increasing.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a transmission unit of a facsimile apparatus which is an example of an image processing apparatus according to the present invention.

FIG. 2 is a block diagram illustrating a configuration of an MTF correction circuit.

FIG. 3 is an explanatory diagram illustrating a configuration example of a filter a.

FIG. 4 is an explanatory diagram showing a configuration example of a filter b.

FIG. 5 is an explanatory diagram showing a relationship between a target pixel and peripheral pixels.

FIG. 6 is an explanatory diagram showing an example of the configuration of a table a.

FIG. 7 is an explanatory diagram illustrating an example of a configuration of a table b.

FIG. 8 is a flowchart illustrating an operation of the MTF correction circuit.

[Explanation of symbols]

 Reference Signs List 10 scanner 20 line buffer 30 MTF correction circuit 40 γ correction section 50 binarization section 60 image memory 70 printer 80 communication section 31a filter a 31b filter b 32 maximum value detection section 33a table a 33b table b 34 selector 35 multiplier circuit 35 addition circuit

Claims (7)

[Claims] 1. A filter unit for calculating an enhancement amount based on a density difference between a target pixel and a peripheral pixel of input image data, and an enhancement coefficient corresponding to a density of the target pixel and a sign of the enhancement amount output by the filter unit. And an enhancement coefficient output unit that outputs a larger enhancement coefficient when the enhancement amount is negative than when the enhancement amount is positive when the density of the pixel of interest is at an intermediate level. A multiplying circuit for multiplying the emphasis amount output by the filter unit by the emphasis coefficient output by the emphasis coefficient output unit; and adding the multiplied value by the multiplying circuit and the density of the pixel of interest to output a corrected density. MT provided with an addition circuit
F correction circuit. 2. The MTF according to claim 1, wherein the filter unit includes a plurality of filters having different characteristics, and a maximum value detection unit that outputs a maximum value of outputs of the filters as an enhancement amount.
Correction circuit. 3. The filter section includes two filters, one filter having a coefficient for enhancing a density difference between a target pixel and a peripheral pixel located at an angle of 45 degrees with respect to the target pixel, and the other filter having a coefficient for enhancing the density difference. 3. The MTF correction circuit according to claim 2, wherein the MTF correction circuit is a filter having a coefficient for enhancing a density difference between a pixel and peripheral pixels on the front, rear, left, and right sides. 4. A first table for setting a coefficient corresponding to each density when the sign of the enhancement amount is positive or zero, and outputting a coefficient corresponding to the density of the pixel of interest, wherein: A second table for setting each coefficient corresponding to each density when the sign of the enhancement amount is negative and outputting a coefficient corresponding to the density of the pixel of interest, and an output of the first table according to the sign of the enhancement amount 4. The MTF correction circuit according to claim 2, further comprising a selector for selecting any one of the output of the second table and the output of the second table. 5. The MTF correction circuit according to claim 4, wherein each coefficient in the first table and the second table is set to have a larger value when the density of the pixel of interest is low. 6. An image reading apparatus for reading an image and outputting digital image data, a line including a pixel of interest of the digital image data output by the image reading apparatus, and a line for holding several lines of digital image data before and after the line. A buffer, a filter unit for calculating an enhancement amount based on a density difference between the pixel of interest and peripheral pixels of the digital image data, and an enhancement coefficient corresponding to the density of the pixel of interest and a sign of the enhancement amount output by the filter unit When the density of the target pixel is at an intermediate level, the enhancement coefficient output unit that outputs a larger enhancement coefficient when the enhancement amount is negative than when the enhancement amount is positive, and the filter unit includes: A multiplication circuit that multiplies the enhancement amount to be output by the enhancement coefficient output by the enhancement coefficient output unit;
And an adding circuit for adding a multiplied value by the multiplying circuit and the density of the target pixel to output a corrected density.
An image processing storage provided with an F correction circuit. 7. A filter section in the MTF correction circuit,
The image processing apparatus according to claim 6, further comprising: a plurality of filters having different characteristics; and a maximum value detection unit that outputs a maximum value of outputs of the filters as an enhancement amount.